C-FAHR Colloquium Series
September 8, 2014
In Winter 2014 –
SVP Ruth Watkins
issued a call for
proposals to a
new program.
Led by Cindy Berg, a team of approximately 30 faculty
members quickly assembled and quickly created a
proposal related to families and health across the
lifespan.
Our proposal was selected as one of the 4
TEP - Cluster Hiring Initiatives .
• Lifespan Health and Wellness
• Digital Humanities
• Large-Scale Data Analysis/Utah Statistical Center
• Society, Water and Climate
Cluster Hiring
Already Filled
• Communicating Complex Health Info
Kim Kapinghst arrives in October 2014
Currently Searching – to begin in July 2015
• Quantitative Methods
• Intervention
Please see website for a flyer to circulate . Search Committee Co-Chairs: Ken Smith &
Jon Butner for Methods position. Robin Marcus & Becky Utz for Intervention position.
Future Possible Positions
• Social Networks
• Dissemination & Implementation
• Family Diversity and Culture
Who are we?
What can we become?
Purpose of C-FAHR
A research consortium to encourage/facilitate interdisciplinary research related to
families and health across the life course.
• How can the family be used as a vehicle for promoting health and adjusting to
chronic illnesses and how such strategies may vary across developmental life
stages (infancy, childhood, adolescence, across adulthood)?
• Multi-level approach (genetic and environmental factors, methodologies,
family interactions and interventions, and healthcare communication) to
understanding health across the lifespan.
http://csbs.utah.edu/health_family.php
Membership
Initial team for the TEP proposal had approximately 30 faculty from 11 different
departments, representing 5 different colleges/institutes
We are now beginning a campus-wide effort to expand
the list of Affiliated Members.
•
•
Submit your biosketch to C-Fahr_info@utah.edu.
Further details on http://csbs.utah.edu/health_family.php
Colloquium Series
Monthly events. To be announced.
Pilot Grant Program
To be announced this fall.
• To support interdisciplinary collaborations among Affiliated Members
• To facilitate pilot work that will aid in the preparation and submission of
external research proposals related to families and health across the life
course.
The Pac-12 Six-Pack
Developing a Comprehensive,
Family-Based Model to Health
Family
Chronic Illnesses are a Family Affair
Berg et al. (2007); Beveridge, Berg, Wiebe & Palmer (2006); King, Berg et al. (2013).
• Family members view type 1 diabetes as a “family”
issue (70.9% mothers, 52.8% children).
• When parents collaborate, are warm and accepting,
and monitor adolescents’ diabetes management,
adherence and HbA1c are improved.
Daily Interdependence of Parent and Child in Managing Diabetes
b=-5.00*
M Persuasiont
BGt+1
BGt
Berg et al. (2013)
Health Psychology
Collaborative Family Involvement Has Benefits and
Liabilities: Couples Coping with Prostate Cancer
Daily collaborative
coping associated
with lower negative
affect and higher
perceived coping
effectiveness for
husbands and wives
Berg et al. (2008)
Psychology and Aging
A potential cost to
collaboration may be
negative affect
transmission
Berg, Wiebe, & Butner
(2011), Gerontology
Family
Genetics and Familiality
Ken Smith, FCS & Population Sciences HCI
No one disputes that genetics matter for psychosocial and
socioeconomic phenomena
Gene-Environment Interaction is key
Familiality-Environment Interaction is even keyer
Yet, several social science disciplines have resisted
introducing genetics, biology, heritability and
familiality into models
PSID, NHANES, ADHEALTH now include genetic data
and family-based designs
2014 NAS & NIA Sponsored “Expert Meeting on
Assessing and Encouraging Interaction between Genetic
and Social-Behavioral Models”
Utah Population Database
Studying Families and Health
• Contributes to numerous groundbreaking genetic and
public health discoveries spanning the past 35 years
• Holds multigenerational pedigrees, 12 generations
deep
• Links family trees to cancer records spanning the past
50 years using the comprehensive Utah Cancer
Registry
• Includes critical birth and death records from the past
century provided by the Utah Department of Health
• Connects to medical records from Intermountain
Healthcare & University of Utah Health Sciences Ctr.
• Allows for recruitment
Brother 105
Sister 98
All Cause Mortality and APOE
(e33 is reference)
P<.05
2
1.8
1.576
1.6
1.4
1.16
1.2
1
0.8
0.892
e23
0.956
e24
e34
e44
Cox regression, includes controls for baseline age, age2, gender, education, LDS affiliation, FEL
19
All Cause Mortality Hazard Rate Ratios for APOE Genotypes (ε33 is
reference)
P<.05
Interaction with Education
Top 87.5% of Education
Bottom 12.5% of Education
3
3
2.6
2.6
2.2
2.2
1.8
1.8
1.4
1.15
1.37
1
2.48
1.31
1.4
1
e34
e44
e34
Interaction between e44 and interval Education p<.05
e44
20
Adversity (Very Young/Old Age)
Among Maternal Grandmothers
Grandmother
Female
Fetus
Germ
Line
(Ego)
Mother
Fetus
(Ego)
Ego
Family
Social Contexts of Health
Ming Wen, Sociology
There have been calls to incorporate the social context
into the study of health. Social context may refer to:
•
•
•
•
•
cultural norms
networks of social relationships
family resources and socialization
neighborhood/community environments
policies and programs
Social contexts, or environments, place constraints on
individual choice. They shape health behaviors by:




shaping norms
enforcing patterns of social control
providing or not providing opportunities to engage in certain behaviors
reducing or producing stress for which certain behaviors may be an
effective coping strategy, at least in the short term.
Social environments vary across the life course.
Example 1: Adolescent Smoking
(Wen, Duker & Olson 2008)
Purpose:
to examine multilevel factors of adolescent smoking after
controlling for the baseline smoking behavior and individual characteristics.
Data: National Longitudinal Study of Adolescent Health (first two waves)
Key findings
 Peer, family and school were all important life domains contextually
influencing smoking behavior among adolescents.
 Time spent with peers, best friend smoking and household member
smoking were associated with higher risk.
 Parent-child closeness, parental control, attending a private school and
having a higher percentage of Hispanic students were protective
factors.
 None of the neighborhood and state-level factors significant in the final
full model (but significant in reduced models).
Example 2: Adult Obesity
(Wen & Kowaleski-Jones 2012)
Purpose:
to examine racial/ethnic disparities in risk of obesity based on
objectively measured body mass and to explore the role of neighborhood built
environment in contributing to these observed disparities.
Data: National Health and Nutrition Examination Survey (2003-2008)
merged with a range of neighborhood data
Key findings
 Neighborhood socioeconomic status and ethnic heterogeneity
negatively linked to risk of obesity controlling for other covariates.
 The built environment is a significant correlate of obesity risk.
Neighborhood walkability (street connectivity and % walking to work),
density, and distance to parks are significant covariates of obesity risks
over and beyond individual risks
 Neighborhood factors were not mediators of racial-ethnic disparities in
obesity.
Family
The Role of Health Care
Joseph Stanford, Family & Preventive Medicine
Do clinicians treat individuals or families?
– Family Medicine, Pediatrics, + other primary care
How can a family perspective inform and
enhance the current transformations in U.S.
health care delivery?
– Accountable care organizations
– Patient-centered medical home
Example: Fertility Decisions
Office of Cooperative Reproductive Health, DFPM, UU
• Fertility decisions and outcomes (family
planning, fertility treatment) are usually
studied from the woman’s perspective.
• Several cohort studies enrolling women and
men with separate longitudinal assessments
of each over time.
Example: Fertility motivations
(Stanford PI, Office of Population Affairs, HHS)
Purpose:
To investigate couple motivations, intentions, behaviors, and
outcomes in family planning
Data: Cohort, multicenter (CEIBA study)
Key finding: Dis/concordance of couple feelings about hypothetical
pregnancy in each cycle
Women
Men
Unhappy
Neutral
Happy
TOTAL
Unhappy
5.4%
2.7%
5.4%
13.4%
Neutral
3.1%
4.0%
11.1%
18.2
Happy
4.8%
6.5%
57.1%
68.3%
TOTAL
13.3%
13.1%
73.6%
100%
Example: Marital fertility
(Stanford and Smith, J Biosoc Sci 2013)
Purpose:
Investigate the association between socioeconomic status,
religion and marital fertility in Utah.
Data: Utah 1996 Health Status Survey
Key finding: Income positively associated with fertility among LDS;
negatively associated among non-LDS.
4
3.5
Children Born
3
2.5
2
1.5
1
0.5
0
<$35,000
$35-55,000
>$55,000
non-lds
2.44047
1.84031
1.39983
inactive lds
1.95233
2.03878
2.07826
active lds
2.53924
3.5008
3.39667
Family
Methods & Data
Methods & Data
Jonathan Butner, Psychology
Family relationships are complex webs of influence
We need different metaphors for understanding
families and health and methods/statistics that enable
those different metaphors
– I use Dynamical Systems Theory
– Multiple outcomes measured repeatedly through time
Two Example Metaphors:
– Family Health as a Coordinated System
– Family Health as a Map
Health as Coordination
Coordination is a taxonomy of how ‘things’ move
together through time
We can build models in Structural Equation Modeling
specifically designed to extract the coordination taxonomy
Added Different Natural
Tendencies for X and Y
In Sync at 3:1 Ratio
A
Anti-Phase 5:1 Ratio, weaker Synchrony
C
B
No Synchrony
D
From ADAPT Study Where Mom/Dad Monitoring are Entrained and
Adolescent Efficacy/Self Control are Synchronized
9
Predicted Average Values
8
7
6
SE
5
BSC
4
MM
3
DM
2
1
0
1
2
3
4
Time
5
6
(Butner, Berg, Baucom, Weibe, In Press
Multivariate Behavioral Research)
Health as a Map
Maps allow us to visualize very complex patterns
of behavior. For example, a pair of Time Series
translates into a trail
We can Use Math/Statistics That Directly Parallel Topographical Features in Maps
Exploratory or Confirmatory
Exploratory Map Based On All Families
We can then examine what changes
the Map giving a very complex
understanding of the family
Probability of Being in Valley (Color Coded)
Estimated via Mixture Modeling
Adolescent Self Control
Family
Methods & Data
Intervention
Dissemination
Intervention & Dissemination
Rebeccca Utz, Sociology
• Does translation or dissemination of health
promotion or health management
programs/concepts differ when using this
comprehensive family-based perspective?
Yes
• Does a comprehensive family-based perspective
improve the success of intervention efforts?
Yes
Example 1: Bereavement
(Utz, Caserta, Lund, 2013 -- R01AG023090-02 from NIA)
Purpose:
To design , implement, and test a
theoretically-based, group-delivered model of
support for recently bereaved spouses.
Data: Randomized intervention design, with longitudinal survey follow-up
Key findings:
Bereavement-related outcomes showed a general trend
of adaptation or recovery following the loss. However, there were no
differences between treatment and control groups
• Need for individually delivered
and/or individually tailored
interventions
• Family-based perspective
Example 2: Adolescent Obesity
(Coffield, Metos, Utz, Waitzman, 2011)
Purpose:
To evaluate the effects of school wellness policies mandated
by the 2004 Child Nutrition and WIC Reauthorization Act on the prevalence of
overweight and obesity among adolescents.
Data:
Utah Population Database, birth certificates and driver license linked
to school policy data for a cohort of children born 1990 to 1992
Key findings:
• Each additional component included in a district's wellness policy was
associated with as much as: 3.2% lower odds in the prevalence of adolescent
overweight (OR = .968; 95% CI = .941-.997). Effect was primarily driven by
nutrition related policies, not physical activity.
• When multivariate regression models controlled for individual, maternal, and
familial characteristics, as well as characteristics of school district
(environenet). The policy effects were dampened, but remained significant.
Comprehensive, Family-Based Model of Health
Family
Intervention
Dissemination
Discussion of this model’s
potential and promise
Methods & Data
Upcoming Events
C-FAHR Symposium
October 6, 4pm to 8pm at the Officer’s Club
Keynote Address by Dr. Rena Repetti, UCLA Department of Psychology.
Followed by a “Research Mixer” and Cocktail Hour where we can begin to
learn about other research and researchers on campus.
October 7, morning to 130pm
small group discussions related to the data/analytic considerations required
for a comprehensive family-based perspective, and brainstorming related to
forming new interdisciplinary collaborations
Submit your biosketch to become an Affiliated Member
C-FAHR-info@utah.edu